Due to the repercussion of falls on both the health and self-sufficiency of older people and on the financial sustainability of healthcare systems, the study of wearable fall detection systems (FDSs) has gained much attention during the last years. The core of a FDS is the algorithm that discriminates falls from conventional Activities of Daily Life (ADLs). This work presents and evaluates a convolutional deep neural network when it is applied to identify fall patterns based on the measurements collected by a transportable tri-axial accelerometer. In contrast with most works in the related literature, the evaluation is performed against a wide set of public data repositories containing the traces obtained from diverse groups of volunteers during the execution of ADLs and mimicked falls. Although the method can yield very good results when it is hyper-parameterized for a certain dataset, the global evaluation with the other repositories highlights the difficulty of extrapolating to other testbeds the network architecture that was configured and optimized for a particular dataset.

A feature extraction algorithm for texture detection oriented to its implementation on embedded electronics based on a Field-Programmable Gate Array (FPGA) is proposed in this paper. Local pre-processing with smart tactile sensors can help to improve dexterity in artificial hands. Simplicity is the goal in order to achieve a hardware-friendly strategy that can be replicated and integrated with another circuitry. This is interesting, considering that tactile sensors are arrays and FPGAs are capable of parallel execution. The proposal was tested with a custom smart tactile sensor mounted on a Cartesian robot to explore different textures. A comparison with a common feature extraction approach based on the fast Fourier Transform (FFT) computation was also made. In addition, the whole procedure is implemented on a System on Chip (SoC) with the feature extraction on the embedded FPGA and a k-means classifier on an ARM core. The proposed algorithm obtains the spatial frequency components of the tactile signal but not their power. Therefore, some information is lost with respect to that provided by the FFT. Nevertheless, an 89.17% accuracy of the proposed algorithm is obtained versus 91.4% with the FFT when twelve different textures are considered, including complex and fabric textures. There is a noticeable saving in power and hardware resources. In addition, since the size of the feature vector is much smaller, data traffic and memory usage is much lower, and the classifier can be simpler.

Low cost devices such as accelerometers, gyroscopes, or force sensing resistors have been exploited to build wearable systems for gait analysis. These sensors are sometimes mounted on a cane if it is required. This paper proposes the use of a tactile sensor on the cane handle to obtain information about gait phases. Moreover, since the sensor is located at the human-device interface, any shaping of the force at this interface that anticipates the movement can be registered. Experimental results from an instrumented cane are presented to illustrate this behavior that can be exploited in applications that involve closed loop control.

The detection of the properties of objects is essential to deal with the manipulation of objects with artificial hands and grippers. In particular, texture detection is a common challenge in robotics. In the quest for smooth and natural manipulation, response times in the order of milliseconds are needed. In a context where the number of sensors and actuators integrated in the system increases, it becomes necessary to pre-process in local electronics. This pre-processing reduces the number of interconnecting wires that hinder movement, and also the computational load and data traffic. This paper proposes the use of the Goertzel algorithm as an alternative to the common use of FFT to obtain the features used to identify a texture. The lower computational cost of the Goertzel algorithm translates into lower resource and power consumption. This lower cost is observed for a limited number of features to be extracted from the raw signal, which can be assumed for an application that seeks to obtain main features of the manipulated object, and not an exhaustive characterisation of the object. This paper shows that a set of 12 textures can be classified with 84.8% accuracy by extracting features that give the signal power for 16 selected frequencies in the spectrum.

This paper presents a procedure for classifying objects based on their compliance with information gathered using tactile sensors. Specifically, smart tactile sensors provide the raw moments of the tactile image when the object is squeezed and desqueezed. A set of simple parameters from moment-versus-time graphs are proposed as features, to build the input vector of a classifier. The extraction of these features was implemented in the field programmable gate array (FPGA) of a system on chip (SoC), while the classifier was implemented in its ARM core. Many different options were realized and analyzed, depending on their complexity and performance in terms of resource usage and accuracy of classification. A classification accuracy of over $94\%$ was achieved for a set of 42 different classes. The proposed approach is intended for developing architectures with preprocessing on the embedded FPGA of smart tactile sensors, to obtain high performance in real-time complex robotic systems.